Implement Modular AutoML System for Model Selection and Tuning

Description:
Design a lightweight AutoML system that automates model selection and hyperparameter tuning for supported algorithms (e.g., Perceptron, KNN, Decision Tree). It should be terminal-based, configurable, and able to produce ranked results with visualizations.

Expected Tasks:

• Create a Python script automl.py inside a new tools/ or automl/ folder.
• Define a consistent interface for models with .fit() and .predict() methods.
• Implement support for ingesting CSV datasets (with pandas).
• Include multiple algorithms from the repo (at least 4).
• Add a configuration for hyperparameter tuning using grid or random search.
• Train each model using cross-validation and evaluate with metrics like accuracy, F1-score, or RMSE.
• Log results to a CSV file and display the best-performing models.
• Use matplotlib or seaborn to generate performance plots.
• Add CLI arguments to control dataset path, model list, metric, and tuning strategy.
• Document setup and usage in the root README.md and add an example dataset.

Stretch Tasks:

• Add time or resource constraints to avoid slow models dominating.
• Support for classification vs regression auto-detection.
• Optional visual dashboard using streamlit or gradio.